Topic: Model-Based Quantitative MRI Meets Machine Learning
Speaker: Professor Gary Zhang, Department of Computer Science, University College London (UCL)
Date and time: 10:00–11:30, September 20
Venue:
Room 103, BME Building (on-site)
Tencent Meeting ID: 340 832 191 (online)
Host: WANG Qian
Abstract:
Magnetic Resonance Imaging (MRI) has established itself as one of the most versatile medical imaging tools in modern healthcare systems. However, conventional MRI is qualitative and produces image contrast that conflates a multitude of underlying properties of the imaged sample. Model-based quantitative MRI offers fundamental advantages over its conventional counterpart but suffers from its own limitations, due to longer acquisition time and processing time. Recent advances in machine learning present major opportunities to address these limitations. In this talk, Prof. Zhang will present several recent developments in this area from his team.
Biography:
Prof. Zhang is a Professor of Computational Imaging in the Department of Computer Science and Centre for Medical Image Computing, University College London (UCL). His core expertise is in medical imaging, medical image analysis, computational modelling, and machine learning. His research interest is in computational imaging approaches for understanding the structure and function of the brain. He has particular expertise in developing quantitative imaging biomarkers for quantifying brain tissue at both the macroscopic and microscopic scales, using diffusion MRI. He is best known for developing DTI-TK, the top-ranked toolkit for spatial normalization of diffusion MRI data, and for developing Neurite Orientation Dispersion and Density Imaging (NODDI), a first-of-its-kind technique for revealing both gray and white matter tissue microstructures in live human subjects.